两层网络中基于随机几何的分簇干扰对齐研究

发布时间：2018-05-07 06:41

本文选题：Femtocell + Macrocell　；参考：《江苏科技大学》2017年硕士论文

【摘要】：近年来,室内语音和数据业务多发,人们对数据服务质量(Quality of Service,QoS)要求越来越高,传统的宏蜂窝(Macrocell)已不能满足当今人们对通信数据业务的需求,如何有效减少通信网络能源消耗、高效利用频谱资源是运营商下一步必须解决的技术问题。与传统的Macrocell网络相比,Femtocell是一种新型的网络模式,具有覆盖范围小、功率低、成低本且用户可自行安装的优点。然而,由于频谱资源的稀缺性,Macrocell和Femtocell必须分享频率带宽,从而导致Macrocell和Femtocell之间跨层、共信道干扰的问题。为了保证在一个Macrocell覆盖范围内多个Femtocell的QoS,必须对如何有效处理Macrocell和Femtocell构成的两层网络中的干扰进行研究。本文对Macrocell和Femtocell构成的两层网络的干扰问题进行分析研究,利用干扰对齐技术的思想通过在发送端设计波束成型矩阵,使每个接收端的干扰对齐到相同子空间中,这样,有用信号就能在无干扰子空间中传输。全文工作概括如下:首先介绍了两层异构网络的背景现状、Femtocell的关键技术和Femtocell网络中存在的干扰,其次介绍了干扰对齐技术和随机几何理论。再次,为了降低密集部署条件下干扰分析的复杂度,本文提出了一种两层网络中基于K-Medoids算法的分簇干扰对齐方法。本文考虑了基于随机几何理论的三维Femtocell网络结构模型,特别是符合三维泊松点过程分布的Femtocell网络模型。以K-Medoids算法为聚类原则对飞蜂窝基站用户对进行分簇,最后利用干扰对齐技术消除每个簇内的同层干扰,减小了干扰管理复杂度,适用于实际的通信系统。最后,为了降低干扰对齐的处理开销和提高系统性能,给出一种基于预编码和用户调度的干扰对齐方案。该方案应用随机几何理论对网络进行建模,再基于信漏噪比(Signal to Leakage plus Noise Ratio,SLNR)算法进行用户调度,得到每个基站的调度用户集,最后,基站基于最大SLNR的干扰对齐算法消除来自其他用户的上行干扰。仿真表明,相比传统的干扰对齐方案所提方案在系统容量上有明显的提升。
[Abstract]:In recent years, there are many indoor voice and data services, and people are demanding more and more high quality of service (QoS). The traditional Macroceller (Macrocell) can not meet the demand of communication data service, so how to reduce the energy consumption of communication network effectively? Efficient use of spectrum resources is the next step operators must solve the technical problems. Compared with the traditional Macrocell network, Femtocell is a new type of network mode, which has the advantages of small coverage, low power, low cost and users can install themselves. However, due to the scarcity of spectrum resources, Macrocell and Femtocell must share frequency bandwidth, which leads to cross-layer and co-channel interference between Macrocell and Femtocell. In order to ensure the QoS of multiple Femtocell within one Macrocell coverage, it is necessary to study how to deal with the interference effectively in the two-layer network composed of Macrocell and Femtocell. In this paper, the interference problem of the two-layer network composed of Macrocell and Femtocell is analyzed and studied. By designing the beamforming matrix at the transmitter, the interference of each receiver can be aligned to the same subspace by using the idea of interference alignment. Useful signals can then be transmitted in a non-interference subspace. The work of this paper is summarized as follows: firstly, the background of two-layer heterogeneous network is introduced. The key technology of Femtocell and the interference in Femtocell network are introduced. Secondly, the interference alignment technique and random geometry theory are introduced. Thirdly, in order to reduce the complexity of interference analysis in dense deployment, a clustering interference alignment method based on K-Medoids algorithm is proposed in this paper. In this paper, three dimensional Femtocell network model based on stochastic geometry theory is considered, especially the Femtocell network model which accords with the distribution of three dimensional Poisson point process. Based on the clustering principle of K-Medoids algorithm, the user pairs of flying cellular base stations are clustered. Finally, the interference alignment technique is used to eliminate the same layer interference in each cluster, which reduces the complexity of interference management and is suitable for practical communication systems. Finally, in order to reduce the processing overhead of interference alignment and improve the performance of the system, a interference alignment scheme based on precoding and user scheduling is proposed. In this scheme, stochastic geometry theory is used to model the network, and then the user scheduling is based on signal to Leakage plus Noise Ratio SLNR algorithm. Finally, the scheduling user set of each base station is obtained. The interference alignment algorithm based on maximum SLNR is used to eliminate the uplink interference from other users. The simulation results show that compared with the traditional interference alignment scheme, the system capacity is improved obviously.
【学位授予单位】：江苏科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5

【参考文献】